diff options
author | Friendika <info@friendika.com> | 2011-10-28 02:50:00 -0700 |
---|---|---|
committer | Friendika <info@friendika.com> | 2011-10-28 02:50:00 -0700 |
commit | e5992003e88545ffe5df0acc57c850733facdd5e (patch) | |
tree | e873335a8e05870e01f5cc8d637611c4ade7c66e /mod | |
parent | e5a7b3e24b5eccf31be7ef5b78580b599a94952f (diff) | |
download | volse-hubzilla-e5992003e88545ffe5df0acc57c850733facdd5e.tar.gz volse-hubzilla-e5992003e88545ffe5df0acc57c850733facdd5e.tar.bz2 volse-hubzilla-e5992003e88545ffe5df0acc57c850733facdd5e.zip |
better handling of remote urls - especially FB opengraph
Diffstat (limited to 'mod')
-rw-r--r-- | mod/parse_url.php | 72 |
1 files changed, 70 insertions, 2 deletions
diff --git a/mod/parse_url.php b/mod/parse_url.php index 7a520ba92..2df9de476 100644 --- a/mod/parse_url.php +++ b/mod/parse_url.php @@ -46,6 +46,7 @@ function parse_url_content(&$a) { killme(); } + if($url && $title && $text) { $text = '<br /><br /><blockquote>' . $text . '</blockquote><br />'; @@ -67,13 +68,26 @@ function parse_url_content(&$a) { killme(); } - logger('parse_url: data: ' . $s, LOGGER_DATA); +// logger('parse_url: data: ' . $s, LOGGER_DATA); if(! $s) { echo sprintf($template,$url,$url,'') . $str_tags; killme(); } + $matches = ''; + $c = preg_match('/\<head(.*?)\>(.*?)\<\/head\>/ism',$s,$matches); + if($c) { +// logger('parse_url: header: ' . $matches[2], LOGGER_DATA); + try { + $domhead = HTML5_Parser::parse($matches[2]); + } catch (DOMException $e) { + logger('scrape_dfrn: parse error: ' . $e); + } + if($domhead) + logger('parsed header'); + } + if(! $title) { if(strpos($s,'<title>')) { $title = substr($s,strpos($s,'<title>')+7,64); @@ -84,10 +98,11 @@ function parse_url_content(&$a) { $config = HTMLPurifier_Config::createDefault(); $config->set('Cache.DefinitionImpl', null); - $purifier = new HTMLPurifier($config); $s = $purifier->purify($s); +// logger('purify_output: ' . $s); + try { $dom = HTML5_Parser::parse($s); } catch (DOMException $e) { @@ -157,10 +172,63 @@ function parse_url_content(&$a) { } } + if(! $text) { + logger('parsing meta'); + $items = $domhead->getElementsByTagName('meta'); + if($items) { + foreach($items as $item) { + $property = $item->getAttribute('property'); + if($property && (stristr($property,':description'))) { + + $text = $item->getAttribute('content'); + if(stristr($text,'<script')) { + $text = ''; + continue; + } + $text = strip_tags($text); + + + $text = substr($text,0,250) . '...' ; + } + if($property && (stristr($property,':image'))) { + + $image = $item->getAttribute('content'); + if(stristr($text,'<script')) { + $image = ''; + continue; + } + $image = strip_tags($image); + + $i = fetch_url($image); + if($i) { + require_once('include/Photo.php'); + $ph = new Photo($i); + if($ph->is_valid()) { + if($ph->getWidth() > 300 || $ph->getHeight() > 300) { + $ph->scaleImage(300); + $new_width = $ph->getWidth(); + $new_height = $ph->getHeight(); + $image = '<br /><br /><img height="' . $new_height . '" width="' . $new_width . '" src="' .$image . '" alt="photo" />'; + } + else + $image = '<br /><br /><img src="' . $image . '" alt="photo" />'; + } + else + $image = ''; + + } + } + } + } + } + if(strlen($text)) { $text = '<br /><br /><blockquote>' . $text . '</blockquote><br />'; } + if($image) { + $text = $image . '<br />' . $text; + } $title = str_replace(array("\r","\n"),array('',''),$title); $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags; |